Transition Metal N-Heterocyclic Carbene (NHC) Complexes in Catalytic Transfer Hydrogenation
Joe DePasquale
12/04/2009
Once thought to be too unstable for practical purposes, carbenes have proved to be a useful class of compounds in coordination chemistry. Carbene complexes contain a carbon atom with a lone pair of electrons and two single bonds, resulting in only six valence electrons around the carbon(1). The lack of an octet of electrons around the carbon atom is where the instability of the complex is derived. In the late twentieth century the development of a relatively stable type of carbene, the N-Heterocyclic Carbene (NHC), enabled the chemical capabilities of carbene complexes to be exploited. Specifically, these NHC complexes have proved useful as stable spectator ligands in transition metal coordination complexes used as catalysts of various reactions.
The unexpected stability of NHC complexes is due to the carbene carbon belonging to an aromatic system that has the carbene carbon directily bonded to two p donor atoms, which in this case are nitrogen atoms. This donation from the filled p orbital of two nitrogen atoms to the vacant p orbital of the carbene carbon enables a resonance structure to be achieved. This resonance structure allows the carbene to have closer to eight valence electrons, enabling the octet rule to be more closely obeyed(1). This resonance and partial multi bond character (displayed as dashed lines) of the C-N-C bonds can be seen in the NHC in Figure 1 below. The p-electron interaction of the donor nitrogen atoms with the porbitals of the carbene carbon raises the energy of these p orbitals, which stabilizes the bent singlet state geometry rather than the linear geometry of the triplet state(1). A thorough explanation of the different geometries of the different states requires complex M.O theory, but the important result is that the singlet state leads to a stronger sigma donor character of the complex, which makes very strong coordinate covalent bonds to transition metals possible.
Figure 1.
Tertiary phosphines have been shown to be successful spectator ligands in transition metal catalysts. N-heterocyclic carbenes (NHC) (Figure 1. Above) are similar to tertiary phosphines because they are both strong sigma donors and can easily form chelates(2). The strong sigma donor quality enables NHC and phosphine ligands to form very strong coordinate covalent bonds with transition metal centers. The disadvantage of tertiary phosphine lignads in catalysis is that they are very air and moisture sensitive and are vulnerable to decomposition in harsh reaction conditions. NHC ligands are expected to be good replacements for phosphines due to their higher stability as a result of even stronger sigma bond donation(3). Also, the ability of NHC ligands to readily form bis and tris chelates like phosphines but with stronger, more rigid backbones, making them less vulnerable to degradation by reductive elimination(4) is another advantage.
NHC ligands are a sterically and electronically tunable ligand class. Un-like tertiary phosphines, NHC ligands do not have the luxury of having a Tolman map to aid in the determination of the electronic and steric effects that the ligand has on the metal coordination complex. The lack of electronic/steric guide such as a Tolaman map is probably due to the use of NHC ligands being a relatively new hot topic and the fact that phosphine ligands are cone shaped while NHC ligands are fan shaped. A bis NHC is shown below in Figure 2. The steric diversity comes from being able to interchange the R1 groups and X linker groups through relatively simple organic synthesis. It has been shown that R1 can be various alkyl or aryl groups(5). The X linker group controls the bite angle of the chelate and can be either alkyl or borane derived(2). The high steric bulk of this system is beneficial for stabilizing a coordinately unsaturated metal complex, which is needed in most catalytic processes. Steric bulk can stabilize complexes with low coordination number and less than eighteen electrons around the metal center because the bulk blocks undesirable species that can normally bond to that metal center and prematurely ending the catalytic cycle. The R2 groups are capable of controlling the electronics of the ligand at the metal based on whether they are electron donating or withdrawing(6). Electron withdrawing groups make the NHC a weaker sigma donor while electron donating groups make the NHC a stronger sigma donor. Figure 2.
The carbenes above are both of the imidazol-2-ylidene type, which is the most common type of NHC, yet other types do exist, such as the triazole-2-ylidene shown in Figure 3 below. These triazole types are not as strong of sigma donors as the imidazole type due to the extra nitrogen, which is more electronegative than carbon, making triazole NHCs closer in strength to tertiary phosphines(6). Imidazole type NHCs are by far the most common for type of NHC ligands and it is no coincidence that the majority of compounds to be describe below are of the this type, but with a few examples of the traizole NHC. Figure 3.
NHC transition metal complexes have been shown to catalyze Heck and related C-C bond formation, transfer hydrogenation, b-alkylation of secondary alcohols with primary alcohols, animation of alcohols, hydrosilylation of alkynes, H/D exchange, alkene diboration, oxidative cleavage of alkenes to aldehydes, and various asymmetric reactions(7),(2). This paper will focus on the NHC/transition metal complexes that have shown to be catalytically active in transfer hydrogenation.
Transfer hydrogenation is the process of reducing molecules by the addition of dihydrogen, typically to a carbonyl group of a ketone. The source of the hydrogen is usually an alcohol, but in some cases it is water, making the process safer and more environmentally friendly than using hydrogen or hydride sources such as LiAlH4 or NaBH48. Scheme 1 below shows the process(8). This process has shown to be achieved catalytically and the main focus of this paper is on how NHC ligands can be used to stabilize transition metal catalysts for this reaction. The mechanism of the transfer of hydrogen is still up to debate with many justifiable theories existing. Some evidence suggests the exact mechanism is dependent upon the nature of the catalyst used.
Scheme_1.JPG
Scheme 1.
The only group eight metal to prove effective for catalyzing transfer hydrogenation with NHC spectator ligands is ruthenium, although many believe iron may work as a metal center with the added benefit that it is inexpensive. Some mono-dentate complexes of the form RuCl2(NHC) (arene) prepared by Cetinkaya et al. have shown to produce aromatic alcohols from aromatic ketones in yields of 78-95%(9).The difference in these complexes is the R groups. By using ether and aryl groups to vary the combination of two different R groups off of both nitrogen atoms on an imidazole, a variety of catalysts were synthesized. The authors were able to demonstrate that electron withdrawing groups (Cl, Br, F) on the ring of the ketone increased the yield of the alcohol while electron donating groups decreased the yield(9). The electron withdrawing groups must alter the electron density around the carbonyl in a way that increases the efficiency of the catalytic cycle.
Currently the best catalytic results for transfer hydrogenation with chelating NHC ligands have been with a Ru pincer complex synthesized by Peris et al(7). This complex is Ru(CNC)(CO)Br2, where CNC is two imidazole carbenes with n-Bu R groups that are bridged by the 2 and 6 positions of a pyridine. This catalyst was prepared by deprotonation of 2,6-bis-(1-n-butylimidazolium-3-yl)pyridine dibromide with NEt3 to induce coordination to [(COD)RuCl2]n and form Ru(CNC)(CO)Br2(10) . The metal bearing a COD (cyclooctadiene) group is a typical starting material for transition metal chemistry because COD is a good leaving group. The appearance of CO in the complex is believed to be from oxidative addition of CH3CHO, which is from the ethanol solvent. NEt3 is a weaker base than typically used for the carbene formation(10).The CNC ligand is tri-dentate with the point of attachment being the traditional C2 position of both NHCs and the nitrogen of the pyridine(10).Hydogen transfer from IPA to aryl and alkyl ketones was catalyzed by Ru(CNC)(CO)Br2 to produce alcohols(7),(10). The highly air stable catalyst was faster at hydrogenating aryl ketones than alkyl. The catalyst also proved to be selective of the polar carbonyl group over C=C. This complex was also successful in catalyzing the oxidation of olefins(10).
The group nine metals, rhodium and iridium, are by far the most commonly used and documented metal centers for these catalysts. Peris et al. reported the first chelating NHC complexes with Rh(11) and went on to test the catalytic ability of these complexes. One of the highest reported catalytic activities of a Rh(III) complex is that synthesized, characterized, and tested by the Peris group(7). The NHC ligand of this complex was the first tripodal bis carbene coordinated to Rh(III) to the date of the article(12). This complex proved to be successful in catalytic transfer hydrogenation for the reduction of C=O and C=N(12). Low catalyst loading of 0.1 mol % still enabled the reduction of aryl and alkyl ketones with isopropyl alcohol as the hydrogen source and KOH as co-catalyst at slightly elevated temperatures(12).
The Crabtree group has also shown that the Ir(III) analogue (Figure 4. Below) of the Rh(III) complexes mentioned in the above paragraph is even more active in the catalysis of hydrogen transfer(2). This work by Crabtree and co-workers demonstrated not only that these NHC ligands can help stabilize transition metal catalysts, but how the steric tunability of these ligands makes them excellent for developing catalysts for specific reactions. The complexes were synthesized from Ir(cod) metal starting material and have one chelating bis imidazole NHC ligand. The linker group was always methylene with the only parameter altered was the R or wingtip group. The complexes were made with R = methyl, n-Bu, i-Pr, neo-pentyl, Bn, and t-Bu(5).
Figure 4.
The best catalytic results of the Ir(III) bis carbene complex (Figure 4. Above) was with the R= neo-pentyl group(5). Most of the complexes were tested for catalytic transfer hydrogenation with benzopheonone as the ketone to be reduced, and IPA as the alcohol with KOH promoter(5). The substrate/catalyst/base reaction ratios were 1000:1:5 (5). In general the primary R groups of Me, nBu, and neopentyl helped stabilize better catalysts than the secondary R group, isopropyl(5). As stated, the neo-pentyl group was the best with 98% conversion and only needing four minutes for complete conversion in comparison to 90 minutes for the other primary catalysts(5). This neo-pentyl Ir complex is one of the most effective catalysts for hydrogen transfer documented as of 2007(2), although catalyst re-cycling proved to be ineffective(5).
Crabtree and co-workers have used the neo-pentyl Ir(III) catalyst describe above along with two other derivatives to hydrogenate aldehyde functional groups with weaker base co-catalyst(13). The weaker base is needed because aldehydes are prone to a deprotonation(13). They tested the catalytic activity of two derivatives. Both had neopentandediyl linkers as oppose to methylene and one of the complexes had 1,2,4-triazole NHC ligands instead of imidazole(13). It was found that in the presence of IPA and K2CO3 co-catalyst the catalyst analogue with an imidazole bis-NHC ligand and the neopentanediyl linker had a higher catalytic activity when reducing all the aldehydes tested, with TOF (turn over frequency) in the range of 240-3000 TON (turn over number) h-1(13). The weak bases, Cs2CO3 and Rb2CO3, proved to be effective at co-catalyzing the reduction of 2-Naphthaldehyde when all three bis-NHC Ir(III) catalysts were used(13). Overall the catalytic activity of the triazole complex was comparable to that of the imidazole complexes(13), which should come as no surprise since tertiary phoshpine auxiliary ligands have already been used by several groups to catalyze transfer hydrogenation and triazoles have comparable sigma donor power to these phosphines.
An Ir(III) NHC complex was synthesized by Yamaguchi and co-workers and it contains the Cp* (Cp = h-5-pentamethylcyclopentadienyl) spectator ligand in addition to the NHC(14). This complex showed high TON and percent completion for the Oppenauer-type oxidation of alcohols(14). This ability of the Ir(Cp*)(NHC)Ln complex to easily catalyze the oxidation of alcohols led Coberan, Peris, and others to look into similar complexes for catalytic transfer hydrogenation, some of which employee non-traditional NHC chelates.
Peris and Coberan synthesized analogues of the IrCp* complex reported by Yamaguchi and co-worker’s complex described above, with an imidazole NHC with R = methyl or n-butyl, and either H, methyl, or Cl being located on the C4 and C5 positions of the imidazole(15). L = Cl for two coordination sites in the complex. Two findings in this study I believe are highly fundamental to catalytic transfer hydrogenation: 1.) These catalysts proved to be efficient without a base as an external co-catalyst and 2.) Evidence of the specific mechanism was documented(15). The purpose of the external base in the catalytic hydrogenation process is to promote the formation of a metal hydride complex thought to be needed for the catalytic cycle(15). This is the first example of base free catalysis which has a significant impact in hydrogenating base sensitive substrates and avoiding un-wanted side-products(15). The fact that the absence of base did not affect any of the parameters used to measure catalytic activity led to a proposed mechanism involving initiation by the oxidative addition of the IPA source to the metal complex through the O-H bond of the alcohol leading to the IPA being oxidized upon release from the catalyst and the formation of a metal hydride complex ready to complete the catalytic cycle(15). Besides a theoretical study that proposes an ionic mechanism with sequential H+ and H- transfer(16) this is one of the few papers that proposes a mechanism for the catalytic hydrogenation process different from the Meerwein-Ponndorf-Verley reduction with main group alkoxides(8). It is also worth noting that this study by Peris and Coberan also gave some asymmetric results showing that the catalyst was selective of alcohol geometry (product).
NHC ligands are useful as spectator ligands for catalysts not only because they are stable, rigid and bulky, but because they can easily be altered by organic synthesis to form a vast assortment of different ligands, which gives hope that some combination can fit the goals of the desired function of the catalyst. Also, the removal of a proton from carbon to form the carbene complex and eventual coordination to the metal is well practiced with many established synthetic routes including: in situ deprotonation of azolium salts with a strong base, transmetallation from a silver-NHC complex, activation of the C2-X bond of an imidazolium cation, plus a few other routes(17).One of the benefits of this structural alteration capability of the NHC ligand is the ability to form chelates. Chelates add stability to the complex via entropy but when all the binding sites are to the metal a disadvantage that occurs is that the metal now has limited coordination sites for binding substrates for activation and forming the necessary intermediates in the cycle. The presence of other bulky ligands in the complex also makes the multi-coordination somewhat of a drawback. These type of issues have led a few research groups to synthesize and employee non-traditional functionalized NHC complexes(18). A study by Peris et al. takes advantage of bis chelating stabilization, but by using a pentamethylcyclopentadienyl-functionalized NHC ligand (Cp*-NHCMe) there by enabling chelation without sacrificing an additional coordination spot(19).
The (Cp*-NHCMe) was synthesized by deprotonation of the methylene group of bezylimidazole with the extremely strong base, n-BuLi(19). This ligand was then coordinated to [IrCl(cod)](2), again taking advantage of the leaving behavior of the COD ligand, to afford the donor functionalized NHC complex shown in Figure 5 below(19). Using the typical alcohol for this type reaction, IPA, the catalytic activity of this complex was tested on aliphatic and aromatic ketones and compared to that of [Cp*IrCl2](2) in order to also gain an understanding of the role of NHC in the catalyst(19). The Ir complexes with NHC ligands showed superior catalytic activity over that with just Cl ligands in terms of TON and percent yield, thus supporting the stabilizing benefits of these ligands(19). What I thought to be an even more significant result is the enhanced activity of the donor functionalized NHC complex over a Cp*Ir(III)NHC complex made and tested by the same group, with the bis NHC being solely coordinated to the metal center, showing that high ligand coordination to the metal center can inhibit the catalytic transfer hydrogenation(19). This catalyst was also active in catalyzing b-alkylation of secondary alcohols with primary alcohols, and animation or primary alcohols(19).
Figure 5.
Another example of a donor functionalized NHC ligand containing complex that is active in catalytic transfer hydrogenation was demonstrated by Hahn et. al. The ligand precursor synthesized is very interesting in structure in that the back end of the imidazole is fused to a benzene ring and it has two different R groups, methyl and 2-propylene(20). Upon NHC coordination to Ir, the alkene R group binds to the metal to complete the bis chelating coordination(20). The tris chelating analogue for this complex was synthesized when the ligand precursor had both R groups being 2-propylene, thus having two non-traditional groups binding to the metal(20). A third Ir complex was synthesized in this series, this one containing a mono-coordinated NHC bonded just through its traditional C2 position, while both R groups are propyl, forming a four coordinate complex (the other two were five coordinate)(20). When catalytic reduction of cyclohexanone to cyclohexanol was studied with an IPA hydrogen source and KOH co-catalyst it was found that the mono-coordinated, four coordinate, NHC containing complex had higher catalytic activity than the chelating five coordinated complexes thus showing how coordinately less saturated complexes can be more catalytically active(20) even when there is sacrifice of stability through chelation.
In summary, NHC ligands are emerging as a highly useful ligand class for stabilizing transition metal catalysts for many reactions including transfer hydrogenation. NHC ligands are very strong sigma donors, even more than their tertiary phosphine counter parts, enabling very strong coordinate covalent bonds to transition metals. NHC ligands can be designed through relatively simple organic synthesis allowing a vast library of ligands with diverse steric, electronic, and coordination properties. The compounds described in this paper demonstrate how these tunable properties can be altered in an attempt to make the complex best serve as a catalyst for transfer hydrogenation. Typically, performing a reaction catalytically as oppose too stoichiometrically, uses less energy and resources while being more environmentally friendly. I anticipate the development and study of NHC ligands for the purpose of stabilizing transition metal catalysts of many reactions, not just transfer hydrogenation, to be an increasingly explored topic in the near future.
Transition Metal N-Heterocyclic Carbene (NHC) Complexes in Catalytic Transfer Hydrogenation
Joe DePasquale
12/04/2009
Once thought to be too unstable for practical purposes, carbenes have proved to be a useful class of compounds in coordination chemistry. Carbene complexes contain a carbon atom with a lone pair of electrons and two single bonds, resulting in only six valence electrons around the carbon(1). The lack of an octet of electrons around the carbon atom is where the instability of the complex is derived. In the late twentieth century the development of a relatively stable type of carbene, the N-Heterocyclic Carbene (NHC), enabled the chemical capabilities of carbene complexes to be exploited. Specifically, these NHC complexes have proved useful as stable spectator ligands in transition metal coordination complexes used as catalysts of various reactions.
The unexpected stability of NHC complexes is due to the carbene carbon belonging to an aromatic system that has the carbene carbon directily bonded to two p donor atoms, which in this case are nitrogen atoms. This donation from the filled p orbital of two nitrogen atoms to the vacant p orbital of the carbene carbon enables a resonance structure to be achieved. This resonance structure allows the carbene to have closer to eight valence electrons, enabling the octet rule to be more closely obeyed(1). This resonance and partial multi bond character (displayed as dashed lines) of the C-N-C bonds can be seen in the NHC in Figure 1 below. The p-electron interaction of the donor nitrogen atoms with the p orbitals of the carbene carbon raises the energy of these p orbitals, which stabilizes the bent singlet state geometry rather than the linear geometry of the triplet state(1). A thorough explanation of the different geometries of the different states requires complex M.O theory, but the important result is that the singlet state leads to a stronger sigma donor character of the complex, which makes very strong coordinate covalent bonds to transition metals possible.
Figure 1.
Tertiary phosphines have been shown to be successful spectator ligands in transition metal catalysts. N-heterocyclic carbenes (NHC) (Figure 1. Above) are similar to tertiary phosphines because they are both strong sigma donors and can easily form chelates(2). The strong sigma donor quality enables NHC and phosphine ligands to form very strong coordinate covalent bonds with transition metal centers. The disadvantage of tertiary phosphine lignads in catalysis is that they are very air and moisture sensitive and are vulnerable to decomposition in harsh reaction conditions. NHC ligands are expected to be good replacements for phosphines due to their higher stability as a result of even stronger sigma bond donation(3). Also, the ability of NHC ligands to readily form bis and tris chelates like phosphines but with stronger, more rigid backbones, making them less vulnerable to degradation by reductive elimination(4) is another advantage.
NHC ligands are a sterically and electronically tunable ligand class. Un-like tertiary phosphines, NHC ligands do not have the luxury of having a Tolman map to aid in the determination of the electronic and steric effects that the ligand has on the metal coordination complex. The lack of electronic/steric guide such as a Tolaman map is probably due to the use of NHC ligands being a relatively new hot topic and the fact that phosphine ligands are cone shaped while NHC ligands are fan shaped. A bis NHC is shown below in Figure 2. The steric diversity comes from being able to interchange the R1 groups and X linker groups through relatively simple organic synthesis. It has been shown that R1 can be various alkyl or aryl groups(5). The X linker group controls the bite angle of the chelate and can be either alkyl or borane derived(2). The high steric bulk of this system is beneficial for stabilizing a coordinately unsaturated metal complex, which is needed in most catalytic processes. Steric bulk can stabilize complexes with low coordination number and less than eighteen electrons around the metal center because the bulk blocks undesirable species that can normally bond to that metal center and prematurely ending the catalytic cycle. The R2 groups are capable of controlling the electronics of the ligand at the metal based on whether they are electron donating or withdrawing(6). Electron withdrawing groups make the NHC a weaker sigma donor while electron donating groups make the NHC a stronger sigma donor.
Figure 2.
The carbenes above are both of the imidazol-2-ylidene type, which is the most common type of NHC, yet other types do exist, such as the triazole-2-ylidene shown in Figure 3 below. These triazole types are not as strong of sigma donors as the imidazole type due to the extra nitrogen, which is more electronegative than carbon, making triazole NHCs closer in strength to tertiary phosphines(6). Imidazole type NHCs are by far the most common for type of NHC ligands and it is no coincidence that the majority of compounds to be describe below are of the this type, but with a few examples of the traizole NHC.
Figure 3.
NHC transition metal complexes have been shown to catalyze Heck and related C-C bond formation, transfer hydrogenation, b-alkylation of secondary alcohols with primary alcohols, animation of alcohols, hydrosilylation of alkynes, H/D exchange, alkene diboration, oxidative cleavage of alkenes to aldehydes, and various asymmetric reactions(7),(2). This paper will focus on the NHC/transition metal complexes that have shown to be catalytically active in transfer hydrogenation.
Transfer hydrogenation is the process of reducing molecules by the addition of dihydrogen, typically to a carbonyl group of a ketone. The source of the hydrogen is usually an alcohol, but in some cases it is water, making the process safer and more environmentally friendly than using hydrogen or hydride sources such as LiAlH4 or NaBH48. Scheme 1 below shows the process(8). This process has shown to be achieved catalytically and the main focus of this paper is on how NHC ligands can be used to stabilize transition metal catalysts for this reaction. The mechanism of the transfer of hydrogen is still up to debate with many justifiable theories existing. Some evidence suggests the exact mechanism is dependent upon the nature of the catalyst used.
Scheme 1.
The only group eight metal to prove effective for catalyzing transfer hydrogenation with NHC spectator ligands is ruthenium, although many believe iron may work as a metal center with the added benefit that it is inexpensive. Some mono-dentate complexes of the form RuCl2(NHC) (arene) prepared by Cetinkaya et al. have shown to produce aromatic alcohols from aromatic ketones in yields of 78-95%(9).The difference in these complexes is the R groups. By using ether and aryl groups to vary the combination of two different R groups off of both nitrogen atoms on an imidazole, a variety of catalysts were synthesized. The authors were able to demonstrate that electron withdrawing groups (Cl, Br, F) on the ring of the ketone increased the yield of the alcohol while electron donating groups decreased the yield(9). The electron withdrawing groups must alter the electron density around the carbonyl in a way that increases the efficiency of the catalytic cycle.
Currently the best catalytic results for transfer hydrogenation with chelating NHC ligands have been with a Ru pincer complex synthesized by Peris et al(7). This complex is Ru(CNC)(CO)Br2, where CNC is two imidazole carbenes with n-Bu R groups that are bridged by the 2 and 6 positions of a pyridine. This catalyst was prepared by deprotonation of 2,6-bis-(1-n-butylimidazolium-3-yl)pyridine dibromide with NEt3 to induce coordination to [(COD)RuCl2]n and form Ru(CNC)(CO)Br2(10) . The metal bearing a COD (cyclooctadiene) group is a typical starting material for transition metal chemistry because COD is a good leaving group. The appearance of CO in the complex is believed to be from oxidative addition of CH3CHO, which is from the ethanol solvent. NEt3 is a weaker base than typically used for the carbene formation(10).The CNC ligand is tri-dentate with the point of attachment being the traditional C2 position of both NHCs and the nitrogen of the pyridine(10).Hydogen transfer from IPA to aryl and alkyl ketones was catalyzed by Ru(CNC)(CO)Br2 to produce alcohols(7),(10). The highly air stable catalyst was faster at hydrogenating aryl ketones than alkyl. The catalyst also proved to be selective of the polar carbonyl group over C=C. This complex was also successful in catalyzing the oxidation of olefins(10).
The group nine metals, rhodium and iridium, are by far the most commonly used and documented metal centers for these catalysts. Peris et al. reported the first chelating NHC complexes with Rh(11) and went on to test the catalytic ability of these complexes. One of the highest reported catalytic activities of a Rh(III) complex is that synthesized, characterized, and tested by the Peris group(7). The NHC ligand of this complex was the first tripodal bis carbene coordinated to Rh(III) to the date of the article(12). This complex proved to be successful in catalytic transfer hydrogenation for the reduction of C=O and C=N(12). Low catalyst loading of 0.1 mol % still enabled the reduction of aryl and alkyl ketones with isopropyl alcohol as the hydrogen source and KOH as co-catalyst at slightly elevated temperatures(12).
The Crabtree group has also shown that the Ir(III) analogue (Figure 4. Below) of the Rh(III) complexes mentioned in the above paragraph is even more active in the catalysis of hydrogen transfer(2). This work by Crabtree and co-workers demonstrated not only that these NHC ligands can help stabilize transition metal catalysts, but how the steric tunability of these ligands makes them excellent for developing catalysts for specific reactions. The complexes were synthesized from Ir(cod) metal starting material and have one chelating bis imidazole NHC ligand. The linker group was always methylene with the only parameter altered was the R or wingtip group. The complexes were made with R = methyl, n-Bu, i-Pr, neo-pentyl, Bn, and t-Bu(5).
Figure 4.
The best catalytic results of the Ir(III) bis carbene complex (Figure 4. Above) was with the R= neo-pentyl group(5). Most of the complexes were tested for catalytic transfer hydrogenation with benzopheonone as the ketone to be reduced, and IPA as the alcohol with KOH promoter(5). The substrate/catalyst/base reaction ratios were 1000:1:5 (5). In general the primary R groups of Me, nBu, and neopentyl helped stabilize better catalysts than the secondary R group, isopropyl(5). As stated, the neo-pentyl group was the best with 98% conversion and only needing four minutes for complete conversion in comparison to 90 minutes for the other primary catalysts(5). This neo-pentyl Ir complex is one of the most effective catalysts for hydrogen transfer documented as of 2007(2), although catalyst re-cycling proved to be ineffective(5).
Crabtree and co-workers have used the neo-pentyl Ir(III) catalyst describe above along with two other derivatives to hydrogenate aldehyde functional groups with weaker base co-catalyst(13). The weaker base is needed because aldehydes are prone to a deprotonation(13). They tested the catalytic activity of two derivatives. Both had neopentandediyl linkers as oppose to methylene and one of the complexes had 1,2,4-triazole NHC ligands instead of imidazole(13). It was found that in the presence of IPA and K2CO3 co-catalyst the catalyst analogue with an imidazole bis-NHC ligand and the neopentanediyl linker had a higher catalytic activity when reducing all the aldehydes tested, with TOF (turn over frequency) in the range of 240-3000 TON (turn over number) h-1(13). The weak bases, Cs2CO3 and Rb2CO3, proved to be effective at co-catalyzing the reduction of 2-Naphthaldehyde when all three bis-NHC Ir(III) catalysts were used(13). Overall the catalytic activity of the triazole complex was comparable to that of the imidazole complexes(13), which should come as no surprise since tertiary phoshpine auxiliary ligands have already been used by several groups to catalyze transfer hydrogenation and triazoles have comparable sigma donor power to these phosphines.
An Ir(III) NHC complex was synthesized by Yamaguchi and co-workers and it contains the Cp* (Cp = h-5-pentamethylcyclopentadienyl) spectator ligand in addition to the NHC(14). This complex showed high TON and percent completion for the Oppenauer-type oxidation of alcohols(14). This ability of the Ir(Cp*)(NHC)Ln complex to easily catalyze the oxidation of alcohols led Coberan, Peris, and others to look into similar complexes for catalytic transfer hydrogenation, some of which employee non-traditional NHC chelates.
Peris and Coberan synthesized analogues of the IrCp* complex reported by Yamaguchi and co-worker’s complex described above, with an imidazole NHC with R = methyl or n-butyl, and either H, methyl, or Cl being located on the C4 and C5 positions of the imidazole(15). L = Cl for two coordination sites in the complex. Two findings in this study I believe are highly fundamental to catalytic transfer hydrogenation: 1.) These catalysts proved to be efficient without a base as an external co-catalyst and 2.) Evidence of the specific mechanism was documented(15). The purpose of the external base in the catalytic hydrogenation process is to promote the formation of a metal hydride complex thought to be needed for the catalytic cycle(15). This is the first example of base free catalysis which has a significant impact in hydrogenating base sensitive substrates and avoiding un-wanted side-products(15). The fact that the absence of base did not affect any of the parameters used to measure catalytic activity led to a proposed mechanism involving initiation by the oxidative addition of the IPA source to the metal complex through the O-H bond of the alcohol leading to the IPA being oxidized upon release from the catalyst and the formation of a metal hydride complex ready to complete the catalytic cycle(15). Besides a theoretical study that proposes an ionic mechanism with sequential H+ and H- transfer(16) this is one of the few papers that proposes a mechanism for the catalytic hydrogenation process different from the Meerwein-Ponndorf-Verley reduction with main group alkoxides(8). It is also worth noting that this study by Peris and Coberan also gave some asymmetric results showing that the catalyst was selective of alcohol geometry (product).
NHC ligands are useful as spectator ligands for catalysts not only because they are stable, rigid and bulky, but because they can easily be altered by organic synthesis to form a vast assortment of different ligands, which gives hope that some combination can fit the goals of the desired function of the catalyst. Also, the removal of a proton from carbon to form the carbene complex and eventual coordination to the metal is well practiced with many established synthetic routes including: in situ deprotonation of azolium salts with a strong base, transmetallation from a silver-NHC complex, activation of the C2-X bond of an imidazolium cation, plus a few other routes(17).One of the benefits of this structural alteration capability of the NHC ligand is the ability to form chelates. Chelates add stability to the complex via entropy but when all the binding sites are to the metal a disadvantage that occurs is that the metal now has limited coordination sites for binding substrates for activation and forming the necessary intermediates in the cycle. The presence of other bulky ligands in the complex also makes the multi-coordination somewhat of a drawback. These type of issues have led a few research groups to synthesize and employee non-traditional functionalized NHC complexes(18). A study by Peris et al. takes advantage of bis chelating stabilization, but by using a pentamethylcyclopentadienyl-functionalized NHC ligand (Cp*-NHCMe) there by enabling chelation without sacrificing an additional coordination spot(19).
The (Cp*-NHCMe) was synthesized by deprotonation of the methylene group of bezylimidazole with the extremely strong base, n-BuLi(19). This ligand was then coordinated to [IrCl(cod)](2), again taking advantage of the leaving behavior of the COD ligand, to afford the donor functionalized NHC complex shown in Figure 5 below(19). Using the typical alcohol for this type reaction, IPA, the catalytic activity of this complex was tested on aliphatic and aromatic ketones and compared to that of [Cp*IrCl2](2) in order to also gain an understanding of the role of NHC in the catalyst(19). The Ir complexes with NHC ligands showed superior catalytic activity over that with just Cl ligands in terms of TON and percent yield, thus supporting the stabilizing benefits of these ligands(19). What I thought to be an even more significant result is the enhanced activity of the donor functionalized NHC complex over a Cp*Ir(III)NHC complex made and tested by the same group, with the bis NHC being solely coordinated to the metal center, showing that high ligand coordination to the metal center can inhibit the catalytic transfer hydrogenation(19). This catalyst was also active in catalyzing b-alkylation of secondary alcohols with primary alcohols, and animation or primary alcohols(19).
Figure 5.
Another example of a donor functionalized NHC ligand containing complex that is active in catalytic transfer hydrogenation was demonstrated by Hahn et. al. The ligand precursor synthesized is very interesting in structure in that the back end of the imidazole is fused to a benzene ring and it has two different R groups, methyl and 2-propylene(20). Upon NHC coordination to Ir, the alkene R group binds to the metal to complete the bis chelating coordination(20). The tris chelating analogue for this complex was synthesized when the ligand precursor had both R groups being 2-propylene, thus having two non-traditional groups binding to the metal(20). A third Ir complex was synthesized in this series, this one containing a mono-coordinated NHC bonded just through its traditional C2 position, while both R groups are propyl, forming a four coordinate complex (the other two were five coordinate)(20). When catalytic reduction of cyclohexanone to cyclohexanol was studied with an IPA hydrogen source and KOH co-catalyst it was found that the mono-coordinated, four coordinate, NHC containing complex had higher catalytic activity than the chelating five coordinated complexes thus showing how coordinately less saturated complexes can be more catalytically active(20) even when there is sacrifice of stability through chelation.
In summary, NHC ligands are emerging as a highly useful ligand class for stabilizing transition metal catalysts for many reactions including transfer hydrogenation. NHC ligands are very strong sigma donors, even more than their tertiary phosphine counter parts, enabling very strong coordinate covalent bonds to transition metals. NHC ligands can be designed through relatively simple organic synthesis allowing a vast library of ligands with diverse steric, electronic, and coordination properties. The compounds described in this paper demonstrate how these tunable properties can be altered in an attempt to make the complex best serve as a catalyst for transfer hydrogenation. Typically, performing a reaction catalytically as oppose too stoichiometrically, uses less energy and resources while being more environmentally friendly. I anticipate the development and study of NHC ligands for the purpose of stabilizing transition metal catalysts of many reactions, not just transfer hydrogenation, to be an increasingly explored topic in the near future.
References
(1) Hahn, F. E.; Jahnke, M. C. Agnew. Chem. , Int. Ed. 2008, 47, 3122-3172. http://dx.doi.org/10.1002/anie.200703883
(2) Mata, J. A.; Poyatos, M.; Peris, E. Coord. Chem. Rev. 2007, 251, 841-859. http://dx.doi.org/doi:10.1016/j.ccr.2006.06.008
(3) Peris, E.; Crabtree, R. H. Coord. Chem. Rev. 2004, 248, 2239-2246. http://dx.doi.org/doi:10.1016/j.ccr.2004.04.014
(4) Crudden, C. M.; Allen, D. P. Coord. Chem. Rev. 2004, 248, 2247-2273. http://dx.doi.org/doi:10.1016/j.ccr.2004.05.013
(5) Albrecht, M.; Miecznikowski, J. R.; Samuel, A.; Faller, J. W.; Crabtree, R. H. Organometallics 2002, 21, 3596-3604. http://dx.doi.org/10.1021/om020338x
(6) Diez-Gonzalez, S.; Nolan, S. P. Coord. Chem. Rev. 2007, 251, 874-883. http://dx.doi.org/10.1016/j.ccr.2006.10.004
(7) Corberan, R.; Mas-Marza, E.; Peris, E. Eur. J. Inorg. Chem. 2009, 2009, 1700-1716. http://dx.doi.org/10.1002/ejic.200801095
(8) Dragutan, V.; Dragutan, I.; Delaube, L.; Demonceau, A. Coord. Chem. Rev. 2007, 251, 765-794. http://dx.doi.org/10.1016/j.ccr.2006.09.002
(9) Ozdemir, I.; Yasar, B.; Cetinkaya, B. Transition Metal Chem. 2005, 30, 831. http://dx.doi.org/10.1007/s11243-005-6736-x
(10) Poyatos, M.; Mata, J. A.; Falomir, E.; Crabtree, R. H.; Peris, E. Organometallics 2003, 22, 1110-1114. http://dx.doi.org/10.1021/om020817w
(11) Poyatos, M.; Sanau, E.; Peris, E. Inorg. Chem. 2003, 42, 2572. http://dx.doi.org/10.1021/ic026212+
(12) Mas-Marza, E.; Poyatos, M.; Sanau, E.; Peris, E. Organometallics 2004, 23, 323-325. http://dx.doi.org/10.1021/om0341552
(13) Miecznikowski, J. R.; Crabtree, R. H. Organometallics 2004, 23, 629-631. http://dx.doi.org/10.1021/om034393x
(14) Hanasaka, F.; Fujita, K.; Yamaguchi, R. Organometallics 2005, 24, 3422-3433. http://dx.doi.org.ezproxy2.library.drexel.edu/10.1021/om0503545
(15) Corberan, R.; Peris, E. Organometallics 2008, 27, 1954-1958. http://dx.doi.org.ezproxy2.library.drexel.edu/10.1021/om800095t
(16) Yamakawa, M.; Ito, H.; Noyori, R. J. Am. Chem. Soc. 2000, 122, 1466-1478. http://dx.doi.org/10.1021/ja991638h
(17) Peris, E. Top. Organomet. Chem. 2007, 21, 83-116. http://showme.physics.drexel.edu/coas/usefulchem/RoutesToNHCs.pdf
(18) Normand, A. T.; Cavell, K. J. Eur. J. Inorg. Chem. 2008, 2008, 2781-2800. http://dx.doi.org/10.1002/ejic.200800323
(19) Pontes da Costa, A.; Viciano, M.; Sanau, M.; Merino, S.; Tejeda, J.; Peris, E.; Royo, B. Organometallics 2008, 27, 1305-1309. http://dx.doi.org/10.1021/om701186u
(20) Hahn, F. E.; Holtgrewe, C.; Pape, T.; Martine, M.; Sola, E.; Oro, L. A. Organometallics 2005, 24, 2203-2209. http://dx.doi.org/10.1021/om0500873